12. Tison F, Dartigues JF, Zuber M, Alperovitch A, Henry P. Prev-
alence of Parkinson’s disease in the elderly: a population study in
Gironde, France. Acta Neurol Scand 1994; 90: 111–15.
13. Baldereschi M, Di Carlo A, Rocca WA et al. for the ILSA
Working Group. Parkinson’s disease and Parkinsonism in a
longitudinal study. Neurology 2000; 55: 1358–63.
14. de Lau LML, Giesbergen PCLM, de Rijk MC, Hofman A,
Koudstaal PJ, Breteler M. Incidence of parkinsonism and Par-
kinson disease in a general population: The Rotterdam Study.
Neurology 2004; 63: 1240–4.
Diuretic use and bone mineral density
in older USA men: the osteoporotic
fractures in men (MrOS) study
SIR—It is estimated that 25–40% of adults aged 65 years
and over use diuretics . The three major classes of diuretics
have different effects on renal calcium balance. Loop diu-
retics increase renal calcium excretion , while thiazide and
potassium-sparing diuretics exert a hypocalciuric effect.
Thiazides facilitate the reabsorption of calcium in the early
segment of the convoluted distal tubule, and potassium-
sparing diuretics promote reabsorption of calcium in the
late segment of the distal tubule [3–5].
Epidemiological data [6–14] and results from ran-
domised controlled trials [15–17] suggest that thiazide
diuretic use is associated with a small increase in bone
mineral density (BMD) or a decreased rate of bone loss.
However, these associations were observed primarily
There are few data regarding the effects of loop diuretics
on BMD and findings are not consistent across studies [18,
19]. Although some studies have shown that loop diuretic
use may be associated with an increased risk of hip and oste-
oporotic fractures [20–22], it is uncertain whether this
possibly elevated fracture risk was mediated by the effect of
loop diuretics on bone or rather on fall-related mechanisms,
such as dizziness and orthostasis. To our knowledge, no
studies have specifically examined whether potassium-
sparing diuretics have an independent effect on BMD.
Therefore, we conducted a cross-sectional study among
participants in the Osteoporotic Fractures in Men (MrOS)
study, an ongoing prospective study of community-dwelling
men, to investigate the association between current diuretic
use and BMD at multiple skeletal sites. We hypothesised
that compared with diuretic non-users, loop diuretic users
would have a lower mean BMD, while thiazide and potassium-
sparing diuretic users would have a higher mean BMD.
The MrOS cohort is comprised of 5,995 men aged 65 years
and older, who were recruited between March 2000 and
April 2002 from six regional areas in the USA: Birmingham,
AL; Minneapolis, MN; Palo Alto, CA; Pittsburgh, PA;
Portland, OR; and San Diego, CA. Written informed
consent was obtained from the participants and the Institu-
tional Review Board at each site approved the study protocol.
During the baseline examination trained interviewers
confirmed participants’ use of diuretics by visual examination
of all current prescription containers. Diuretics were classi-
fied as loop, thiazide, potassium-sparing and thiazide/
potassium-sparing combinations (e.g. Dyazide®, Maxzide®
and Moduretic®). We excluded 40 participants who were
using loop and thiazide diuretics (n=23), loop and
potassium-sparing diuretics (n=14) and all three diuretic
combinations (n=3). Baseline BMD measurements were
obtained at the total hip and its subregions, femoral neck
and trochanter; lumbar spine, and total body using dual X-ray
absorptiometry (QDR 4500W, Hologic, Inc., Waltham, MA).
A baseline questionnaire was used to obtain information
on demographic characteristics (age and race), lifestyle
factors (physical activity, tobacco use and diet) and medical
history. Physical activity was measured by computing the
Physical Activity Scale for the Elderly score based on the
activity level documented by the participant . With
respect to tobacco use, participants were categorised into
past, current and never smokers. Dietary information was
acquired using a modified version of the Block Food
Frequency Questionnaire . Calcium intake was estimated
by summing the average daily consumption of calcium from
diet and supplementation. Potassium intake was the esti-
mated daily consumption of dietary potassium only. Partici-
pants were questioned regarding whether they had diabetes
mellitus, hypertension, congestive heart failure (CHF), chronic
obstructive pulmonary disease (COPD), stroke, angina or
myocardial infarction. Cardiovascular disease was defined as
the presence of any of the latter three conditions. Partici-
pants had their body mass index (BMI) calculated from
their measured weight, in kilograms, divided by the square
of their height in metres (kg/m2).
Differences in baseline characteristics were assessed
using ANOVA for continuous variables and chi-squared
analyses for categorical variables. From among baseline
variables with differences (P<0.10) between diuretic
groups, we selected a final set of variables to include in
the multivariate models. We used ANCOVA to assess
the age- and multivariate-adjusted associations between
each diuretic user group and BMD. Any overall differ-
ence in BMD between categories of diuretic users was
tested using the omnibus F-test with statistical signifi-
cance established at P<0.05. Differences between pairs
of diuretic categories were then tested using the two-
tailed tests of the least squares means. Pairwise compari-
sons were adjusted using the Bonferroni method .
Statistical analyses were performed using PC SAS version
8.2 (SAS Institute, Cary, NC).
Our analyses involved 5,955 men (90% white) with a mean age
of 73 (±5.9) years. There were significant differences among
the different diuretic groups for all selected characteristics
with the exception of dietary calcium and potassium intake
by guest on January 6, 2012
Compared with diuretic non-users, loop diuretic users
had a 2.9–4.6% higher age-adjusted mean BMD at the total
hip, femoral neck and total body (P<0.05 for all comparisons)
(Table 2). Thiazide users had a 2–3.6% greater age-adjusted
mean BMD compared with diuretic non-users at all meas-
ured skeletal sites (P<0.05 for all comparisons). However,
multivariate adjustment diminished the average differences
in BMD observed between specific classes of diuretic users
and non-users. There were no statistically significant differ-
ences in mean BMD between potassium-sparing or thiazide/
potassium-sparing diuretic users and diuretic non-users in
both age- and multivariate-adjusted models after Bonferroni
correction (Table 2). No significant differences in BMD
were observed between categories of diuretic users.
Results of this cross-sectional study suggest that recent diu-
retic use is not associated with BMD in older men. We
observed that the higher age-adjusted mean BMD among
diuretic users was largely accounted for by differences
between participants in prevalent medical conditions and
Despite the previously hypothesised hypocalciuric effect
of loop diuretics, we found no evidence of reduced BMD
among users of loop diuretics. A previous cross-sectional
study of older adults has reported no association between
loop diuretics and ultrasound measurements of the
calcaneus . Another cross-sectional study in postmeno-
pausal women showed that compared with loop diuretic
non-users, users of loop diuretics had a 5% lower BMD at
the femoral neck and trochanter . However, these
results have not been replicated and, to our knowledge,
other studies exploring a relationship between loop diuretic
use and BMD in men have not been reported in the medical
We found that thiazide diuretic users had a significantly
higher age-adjusted mean BMD at all five measured skeletal
sites that diminished with multivariate adjustment. Previ-
ously published data in both men and women have generally
shown significant age-adjusted differences that were attenuated
with increased accounting for possible confounders [6, 8–11].
Table 1. Baseline characteristics among USA male diuretic users and non-users aged 65 years and older: the MrOS
. . . . .. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . .
Number (%)250 (4.3) 100 (1.7)
Mean age (years) 77 75
Mean BMI (kg/m2) 29.2 27.9
Diabetes mellitus (%) 27.6 11.0
Hypertension (%) 67.6 87.0
COPD (%) 20.0 8.0
Cardiovascular disease (%)a 58.0 29.0
CHF (%) 40.8 6.0
Smokers: past (%) 65.6 64.0
Smokers: current (%) 2.8 0
PASEb Score 124 128
Mean daily calcium intake (mg)c 1075 1188
Mean daily potassium intake (mg)d 2887 3055
aHistory of myocardial infarction, angina, or stroke.
bPhysical Activity Scale for the Elderly .
cCalcium from diet and supplements.
dPotassium intake from dietary sources only.
. . . . . . . . . . . . . . . . . .
. . . . .. . . . . . . . . . . .
. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . .. . . . . . . . . . . . . .
Table 2. Percentage difference in BMD (95% CI) between participants using specific classes of diuretics and non-users of
diuretics: the MrOS Study, 2000–2002a
Diuretic group (n) Total hip
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . .
MV-adjustedc % (95% CI)
−0.3 (−2.1, 1.6)
MV-adjustedc % (95% CI) 1.0 (−0.3, 2.3)
MV-adjustedc % (95% CI) 1.9 (−0.8, 4.7)
aPositive values indicate that BMD for the specific diuretic group is greater than that for the comparison group (diuretic non-users, n=4,939).
bP<0.05, Bonferroni adjusted.
cMV=multivariate adjusted for age, body mass index, physical activity, diabetes mellitus, cigarette smoking, congestive heart failure, hypertension, chronic
obstructive pulmonary disease, and cardiovascular disease (myocardial infarction, angina, or stroke).
4.6 (2.6, 6.7)b
0.9 (−1.2, 3.0)
2.9 (1.5, 4.3)b
0.8 (−0.6, 2.3)
3.8 (0.6, 7.0)
2.5 (−0.6, 5.5)
2.0 (−1.0, 4.9)
0.2 (−2.6, 3.1)
1.1 (−1.0, 3.2)
−1.3 (−3.5, 0.9)
2.3 (0.9, 3.8)b
0.8 (−0.7, 2.4)
2.3 (−1.0, 5.6)
1.3 (−1.9, 4.5)
1.0 (−2.0, 4.1)
−0.3 (−3.2, 2.7)
3.0 (0.8, 5.3)
0.4 (−1.9, 2.8)
3.6 (2.0, 5.1)b
1.3 (−0.3, 2.9)
3.3 (−0.1, 6.8)
1.5 (−1.9, 4.9)
0.3 (−2.9, 3.5)
−1.6 (−4.8, 1.5)
2.9 (1.5, 4.3)b
1.6 (0.2, 3.1)
2.0 (1.1, 3.0)b
1.3 (0.3, 2.4)
2.3 (0.2, 4.5)
1.9 (−0.2, 4.0)
0.7 (−1.3, 2.7)
0.2 (−1.8, 2.2)
Age-adjusted % (95% CI) 3.1 (1.3, 5.0)b
Age-adjusted % (95% CI) 2.8 (1.5, 4.1)b
Age-adjusted % (95% CI) 3.1 (0.2, 6.0)
Age-adjusted % (95% CI)
MV-adjustedc % (95% CI)
1.6 (−1.0, 4.3)
0.1 (−2.4, 2.6)
by guest on January 6, 2012
A previous cross-sectional study involving 681 older men
showed that thiazides had no effect on BMD . Although
randomised controlled studies have demonstrated a positive
relationship between thiazide use and BMD [15–17], most
have been performed in women. In one randomised
controlled study, men who received hydrochlorothiazide
did not have significant increases in BMD and no dose–
response relationship was observed .
To our knowledge, there have been no prior studies that
specifically evaluated the use of potassium-sparing diuretics
on BMD. In both age- and multivariate-adjusted models, we
observed that neither potassium-sparing nor combination
thiazide/potassium-sparing diuretics had any significant
influence on bone density in our cohort of older men.
Strengths of this study include its comprehensive evalua-
tion of diuretic use and BMD in a large cohort of older men,
and its examination of the associations of loop, thiazide,
potassium-sparing and combination diuretics with BMD at
multiple skeletal sites. We also made adjustments for the
presence of multiple potentially confounding factors.
The present study also has several limitations. Firstly,
because analyses were cross-sectional and no data are avail-
able on duration of diuretic usage, the temporality between
diuretic use and BMD changes cannot be determined and
causality cannot be inferred. Secondly, the lack of informa-
tion on diuretic dose precluded an examination of dose–
response relationships between the various diuretics and
BMD. Thirdly, multivariate analyses did not account for
potentially important interactions between individual cov-
ariates and BMD. For example, congestive heart failure may
impact body mass index due to fluid retention and sarcopenia;
and cardiovascular disease may impact exercise levels due to
physical impairment or exercise intolerance. We lacked the
data to match subjects for heart disease severity to run a
stratified analysis. Finally, these results may not be generali-
seable to a non-white or younger population of men.
In this cross-sectional study among community-dwelling
men aged 65 years and older, current users of loop, thiazide,
potassium-sparing and thiazide/potassium-sparing diuretics
had a higher age-adjusted mean BMD compared with non-users.
However, after multivariate adjustment, the magnitude of
this difference was diminished and no longer significant.
There were no significant differences in BMD between the
four diuretic groups. Overall, the results suggest that
current diuretic use is not associated with BMD in older,
• Diuretics have varying effects on renal calcium balance
which could influence BMD.
• Although thiazide diuretics have been associated with
modest increases in BMD, not much is known about
the effects of loop, potassium-sparing and thiazide/
potassium-sparing combination diuretics on BMD,
particularly in older men.
• This cross-sectional study demonstrated that, after
adjusting for confounding factors, no significant differ-
ences were found in BMD among men who used or did
not use various diuretics.
Sources of research funding
The Osteoporotic Fractures in Men (MrOS) Study is
supported by National Institutes of Health funding. The
following institutes provide support: the National Institute of
Arthritis and Musculoskeletal and Skin Diseases(NIAMS),
the National Institute on Aging (NIA), and the National
Cancer Institute (NCI), under the following grant numbers:
UO1 AG18197–02, UO1AR45580–02, UO1 AR45614, UO1
AR45632, UO1 AR45647, UO1 AR45654, UO1 AR45583
and M01 RR00334.
LIONEL S. LIM1*, HOWARD A. FINK2, MICHAEL A. KUSKOWSKI2,
JANE A. CAULEY3, KRISTINE E. ENSRUD2, FOR THE OSTEOPOROTIC
FRACTURES IN MEN (MROS) STUDY GROUP
1Erlanger Bledsoe Internal and Geriatric Medicine, PO Box 1108,
Pikeville, TN 37367, USA (work done at Division of Preventive
and Occupational Medicine, Mayo Clinic College of Medicine,
Rochester, MN 55905, USA)
Fax: (+1) 423 447 3621
2Geriatric Research Education & Clinical Center, VA Medical
Center, One Veterans Drive, Minneapolis, MN 55417, USA
3University of Pittsburgh, 130 DeSoto Street, A524, Pittsburgh,
PA 15261, USA
*To whom correspondence should be addressed
1. van Kraaij DJ, Jansen RW, Gribnau FW, Hoefnagels WH.
Loop diuretics in patients aged 75 years or older: general
practitioners’ assessment of indications and possibilities for
withdrawal. Eur J Clin Pharmacol 1998; 54: 323–7.
2. Rose BD. Diuretics. Kidney Int 1991; 39: 336–52.
3. Costanzo LS. Localization of diuretic action in microperfused
rat distal tubules: Ca and Na transport. Am J Physiol 1985;
4. Shimizu T, Nakamura M, Yoshitomi K, Imai M. Interaction of
trichlormethiazide or amiloride with PTH in stimulating Ca2+
absorption in rabbit CNT. Am J Physiol 1991; 261: F36–43.
5. Friedman PA, Gesek FA, Shimizu T et al. Stimulation of
calcium transport by amiloride in mouse distal convoluted
tubule cells. Kidney Int 1995; 48: 1427–34.
6. Morton DJ, Barrett-Connor EL, Edelstein SL. Thiazides and
bone mineral density in elderly men and women. Am J Epidemiol
1994; 139: 1107–15.
7. Wasnich RD, Benfante RJ, Yano K, Heilbrun L, Vogel JM.
Thiazide effect on the mineral content of bone. N Engl J Med
1983; 309: 344–7.
8. Bauer DC, Browner WS, Cauley JA et al. Factors associated
with appendicular bone mass in older women. The Study of
Osteoporotic Fractures Research Group. Ann Intern Med
1993; 118: 657–65.
9. Glynn NW, Meilahn EN, Charron M, Anderson SJ, Kuller
LH, Cauley JA. Determinants of bone mineral density in older
men. J Bone Min Res 1995; 10: 1769–77.
10. Sigurdsson G, Franzson L. Increased bone mineral density in a
population-based group of 70-year-old women on thiazide
diuretics, independent of parathyroid hormone levels. J Intern
Med 2001; 250: 51–6.
by guest on January 6, 2012
11. Cauley JA, Cummings SR, Seeley DG et al. Effects of thiazide
diuretic therapy on bone mass, fractures, and falls. The Study
of Osteoporotic Fractures Research Group. Ann Intern Med
1993; 118: 666–73.
12. Wasnich RD, Ross PD, Heilbrun LK, Vogel JM, Yano K,
Benfante RJ. Differential effects of thiazide and estrogen upon
bone mineral content and fracture prevalence. Obstet Gynecol
1986; 67: 457–62.
13. Wasnich R, Davis J, Ross P, Vogel J. Effect of thiazide on
rates of bone mineral loss: a longitudinal study. BMJ 1990;
14. Dawson-Hughes B, Harris S. Thiazides and seasonal bone
change in healthy postmenopausal women. Bone Minerals
1993; 21: 41–51.
15. Wasnich RD, Davis JW, He YF, Petrovich H, Ross PD. A ran-
domized, double-masked, placebo-controlled trial of chlortha-
lidone and bone loss in elderly women. Osteoporos Int 1995;
16. LaCroix AZ, Ott SM, Ichikawa L, Scholes D, Barlow WE.
Low-dose hydrochlorothiazide and preservation of bone
mineral density in older adults. A randomized, double-
blind, placebo-controlled trial. Ann Intern Med 2000; 133:
17. Reid IR, Ames RW, Orr-Walker BJ et al. Hydrochlorothiazide
reduces loss of cortical bone in normal postmenopausal
women: a randomized controlled trial. Am J Med 2000; 109:
18. van Daele PL, Burger H, Algra D et al. Age-associated
changes in ultrasound measurements of the calcaneus in men
and women: the Rotterdam Study. J Bone Min Res 1994; 9:
19. Ooms ME, Lips P, Van Lingen A, Valkenburg HA. Deter-
minants of bone mineral density and risk factors for oste-
oporosis in healthy elderly women. J Bone Min Res 1993; 8:
20. Tromp AM, Ooms ME, Popp-Snijders C, Roos JC, Lips P.
Predictors of fractures in elderly women. Osteoporos Int
2000; 11: 134–40.
21. Heidrich FE, Stergachis A, Gross KM. Diuretic drug use and
the risk for hip fracture. Ann Intern Med 1991; 115: 1–6.
22. Taggart HM. Do drugs affect the risk of hip fracture in elderly
women? J Am Geriatr Soc 1988; 36: 1006–10.
23. Washburn RA, McAuley E, Katula J, Mihalko SL, Boileau RA.
The physical activity scale for the elderly (PASE): evidence for
validity. J Clin Epidemiol 1999; 52: 643–51.
24. Block G, Woods M, Potosky A, Clifford C. Validation of a
self-administered diet history questionnaire using multiple diet
records. J Clin Epidemiol 1990; 43: 1327–35.
25. Bland JM. Multiple significance tests: the Bonferroni method.
BMJ 1995; 310: 170.
Second hip fracture in elderly hip fracture
patients: cost and effectiveness of fracture
SIR—Hip fractures account for about 10% of all fractures
, increasing by 1–3% annually . About 8% of these
patients sustain repeat hip fracture within the first year .
The probability of sustaining two hip fractures in the course
of an individual’s life could reach 20% . Treatment with
vitamin D and calcium supplements and improvement of
vitamin D status play an important role in the osteoporosis
therapeutic strategy and in the prevention of hip fractures in
the elderly, decreasing the hip fracture rate by 30% in
treated versus untreated patients . Efficient fracture pre-
vention treatment with alendronate has been reimbursed in
Israel since January 2000.
We assessed standards of care, following an index hip
fracture, and the rate of second hip fractures in elderly
patients treated in the community clinics (community-
treated patients: CTP) and compared it with the rate in the
participants of a post-surgical osteoporosis treatment pro-
Participation in the PSOTP was offered to all elderly hip
fracture patients who underwent surgical correction of hip
fracture in the Department of Orthopaedic Surgery during
2001–2002. Frequent fallers, patients having major psychi-
atric problems, malnutrition or active malignant disease in
the last 5 years and patients receiving bisphosphonates for
established osteoporosis were excluded.
Information about the first hip fracture, laboratory evalua-
tion, concomitant diseases and medications was retrieved
from the patients’ discharge charts. Two years after the
index fracture, the CTP were contacted by a physician from
the Metabolic Bone Diseases Unit. The patients were
requested to bring to the phone all their current medications
and to read their names to the interviewer. Questions about
calcium and vitamin D doses were specifically stressed. The
patients were asked about new fractures and about past and
current use of antiresorbing drugs. All reported second hip
fractures were validated by data retrieval from patients’ hos-
The PSOTP patients underwent quarterly clinical and labo-
ratory evaluation. For details please see Appendix 1 in the
supplementary data on the journal website (www.ageing.
oxfordjournals.org). All the PSOTP patients were
treated with 600mg of elemental calcium and 800IU of
vitamin D3. Treatment with alendronate was started after
therapeutic correction of vitamin deficiency (serum
Descriptive statistics were used to characterise the
patients’ groups. Comparison in fracture rates was per-
formed using Fisher’s exact test and the calculation of rela-
tive risk (RR) with 95% confidence interval (CI). To
compare two groups, t-test or Mann–Whitney test (as
appropriate) was used for continuous variables, and chi-
squared test for categorical variables.
Of 512 elderly hip fracture patients who met inclusion crite-
ria, 97 (19%) consented to participate in the PSOTP and
415 (81%) were treated in their community medical care
facilities. Two years after the index fracture, 29 (5.6%) were
by guest on January 6, 2012